Differential Occupation of Axial Morphospace

Home , Callechelys, Delma, Sarcopterygii, Terrestrial locomotion

Zoology 117 (2014) 70–76

Contents lists available at ScienceDirect

Zoology

jou rnal homepage: www.elsevier.com/locate/zool

ଝ

Differential occupation of axial morphospace

a,∗ b

Andrea B. Ward , Rita S. Mehta

Department of Biology, Adelphi University, 1 South Avenue, Garden City, NY 11530, USA

Long Marine Lab, Ecology and Evolutionary Biology, University of California, Santa Cruz, CA 95060, USA

r t a b

i s

c l e i n f o t r a c t

Article history: The postcranial system is composed of the axial and appendicular skeletons. The axial skeleton, which

Received 26 August 2013

consists of serially repeating segments commonly known as vertebrae, protects and provides leverage

Received in revised form 7 October 2013

for movement of the body. Across the vertebral column, much numerical and morphological diversity

Accepted 9 October 2013

can be observed, which is associated with axial regionalization. The present article discusses this basic

Available online 5 December 2013

diversity and the early developmental mechanisms that guide vertebral formation and regionalization.

An examination of vertebral numbers across the major vertebrate clades ﬁnds that actinopterygian and

Keywords:

chondrichthyan ﬁshes tend to increase vertebral number in the caudal region whereas Sarcopterygii

Axial skeleton

increase the number of vertebrae in the precaudal region, although exceptions to each trend exist. Given

Axial regionality

Vertebrae the different regions of axial morphospace that are occupied by these groups, differential developmental

Vertebral development processes control the axial patterning of actinopterygian and sarcopterygian species. It is possible that,

among a variety of factors, the differential selective regimes for aquatic versus terrestrial locomotion

have led to the differential use of axial morphospace in vertebrates.

1. Introduction performance (Tytell et al., 2010). Both vertebral number and

shape of the individual vertebrae have been shown to be useful in

One of the major deﬁning characteristics of Vertebrata is having understanding the evolution of body shape variation within (Ward

a vertebral column that is composed of serially repeating ossi- and Brainerd, 2007; Bergmann and Irschick, 2012) and across

ﬁed, cartilaginous, and ligamentous elements that surround the vertebrate groups (Collar et al., 2013). Furthermore, movement

spinal cord and notochord (Schultze and Arratia, 1988; Janvier, of individual vertebrae within the vertebral column has provided

1997). These repeating structures are known as vertebrae. Over insight into diverse behavior patterns of vertebrates (Moon,

evolutionary history, vertebrae have taken on many functional 1999). Despite these recent studies, the functional implications of

roles including protection of the spinal cord and dorsal aorta vertebral variation still need to be explored.

as well as providing attachment points for the axial muscula- In the present article, our aim is to provide a brief overview

ture. These functional differences correspond to the structural and of the morphological diversity of vertebrae and the early develop-

numerical diversity that we observe across the vertebrate axial mental mechanisms that guide vertebral formation. While doing

skeleton. so, we discuss differentiation of the axial skeleton and the mech-

For centuries, vertebrae were and continue to be a potent anisms underlying regionalization. We also search for broad-scale

morphological character for taxonomic and comparative studies. patterns in the axial skeleton by assembling a large data set con-

Since the 1800s, studies have pointed out the effects of envi- sisting of vertebral numbers in the precaudal and caudal regions

ronmental variation on vertebral characteristics (Jordan, 1891; from diverse members of the major vertebrate clades. Lastly, we

Lindsey, 1975; McDowall, 2008 and references therein). However, highlight exciting avenues for future research.

only during the last ﬁfty years have vertebrae become the focus

of functional biology research. For example, vertebral number has

2. Postcranial axial skeleton and variation in vertebral

provided insight into maximum body length and shape in ﬁshes articulation

(Lindsey, 1975), which, in turn, has been shown to have functional

consequences on important survival traits such as swimming

Although often overshadowed by attention given to the skull,

the postcranial system, composed of the axial and appendicular

skeletons, provides the skull with movement, suspends the body,

ଝ and propels the body through different media. In Chordata, the

This article is part of a special issue entitled “Axial systems and their actuation:

notochord is the primary axial structure. In Agnatha and some

new twists on the ancient body of craniates”.

∗

early Actinopterygii, the notochord continues to be the primary

Corresponding author. Tel.: +1 516 877 4204.

E-mail address: [email protected] (A.B. Ward). axial structure in embryos and adults. It is not until the arrival of

A.B. Ward, R.S. Mehta / Zoology 117 (2014) 70–76 71

the Gnathostomata that the notochord is replaced by a vertebral

column with centra in adults.

On examining the axial skeleton, the morphological diversity of

individual vertebrae is apparent. Vertebrae are composed of multi-

ple components: centra, arches, and intervertebral discs (Schultze

and Arratia, 1988). Any of these components can contribute to the

morphological and functional diversity of the axial skeleton. Verte-

brae have also undergone enlargements, reductions, and fusions,

which have added to the challenge of tracking development and

evolution.

The centrum is the main body of the vertebra; it surrounds

and replaces or incorporates the notochord. Vertebrae vary in type

based on the relationship of the arches to the centrum, the num-

ber of elements forming each centrum, and the embryonic origin

of the centrum. Based on embryonic origin, centra may be cat-

egorized into four different types: chordacentra (when present

as mineralized or calciﬁed rings within the notochordal sheath

in actinopterygians), arcocentra (when formed by ossiﬁcation of

cartilage extending from the arches around the notochord), auto-

centra (when formed independent of the dorsal and ventral arches),

or holocentra (when formed by proliferation of cartilage cells

around the notochord that then ossify to form amphicoelous cen-

tra) (Gadow, 1933; Arratia et al., 2001).

Multiple centra articulate to form the axial column. The artic-

ulating surfaces of the centra vary and this variation can affect

the movement and the roles of the axial skeleton. Centra with

ﬂat ends are termed acoelous whereas centra with concave ends

are amphicoelous. Acoelous centra are found in precaudal verte-

brae, i.e., mainly the trunk, of early reptiles, birds, and mammals.

Amphicoelous centra are found in Chondrichthyes, Actinoptery-

gii, and some amphibians, such as the salamander Necturus. Centra

that are concave anteriorly and convex posteriorly are called pro-

coelous while the reverse of this design is known as opisthocoelous.

In the procoelous and opisthocoelous conditions, the centra artic-

ulate like a ball-and-socket joint and rotation is at the center of

Fig. 1. Regionalization of the vertebral column in Vertebrata. While the vertebral

the joint, reducing the possibility of vertebral dislocation. Exam-

columns of Chondrichthyes and Actinopterygii are divided into a precaudal and

ples of vertebrae with procoelus and opisthocoelous centra can be

caudal region, all other groups have more than two regions. In amniotes, regional-

found in Lissamphibia and early reptiles. Lastly, heterocoelous cen- ization seems to be based upon the presence or absence of ribs or the variation in

tra, which are found in the cervical region of birds and turtles, have ribs along the vertebral column. (A) Mammals have the greatest diversity across the

axial skeleton with ﬁve distinct regions. Also, despite elongation in speciﬁc regions,

saddle-shaped articular surfaces at both ends (Liem et al., 2001).

mammals still maintain 7 cervical vertebrae and 19–20 thoracolumbar vertebrae.

Projections known as apophyses may also be found extending

Mammalian precaudal elongation is mostly due to elongation of the individual cen-

from the centra and vertebral arches. Depending on their size and

tra. (B) Sauropsids have four distinct regions with the thoracolumbar region (also

location, apophyses can articulate with other bones, such as ribs, known as abdominal or dorsal) having the greatest number of vertebrae. (C) A bony

ﬁsh skeleton illustrating the simple divisions of the axial skeleton with the precau-

or form interlocking processes, known as zygopophyses, between

dal and caudal region. Vertebrae in the caudal region are distinguished from those

vertebrae. In some elongate terrestrial amniotes where torsion of

in the precaudal region by the presence of hemal arches.

the body may be extreme, such as snakes, additional zygopophyses,

Line drawings are modiﬁed from Parker (1900), Ellenberger et al. (1956), and Romer

called zygosphenes and zygantra, may be present on the anterior (1966).

and posterior margins, respectively, of the neural spine. It was pro-

posed that torsion at the vertebral joints was not possible due to the

regionalization across vertebrates (Fig. 1). The vertebral column

zygosphene–zygantrum articulations (Mosauer, 1932). However,

may be divided into 2–5 regions depending on the taxon; these

in snakes, ﬂexion of the vertebral column is extremely important for

regional designations are based on anatomical features of the verte-

many survival behaviors such as locomotion, defensive behaviors,

brae as well as their relative placement in the body (Goodrich,

and feeding (see Moon, 1999 for a discussion of axial torsion in dif-

1958). In Chondrichthyes and Actinopterygii, axial regions are fre-

ferent snake taxa during diverse behaviors). Despite the additional

quently divided in two: precaudal and caudal (exceptions will be

morphological restrictions imposed by zygosphene–zygantrum

discussed below). In Tetrapoda, the axial skeleton may contain up

articulations, considerable torsion is possible in the snake axial

to ﬁve distinct regions: cervical, thoracic, lumbar, sacral, and caudal

skeleton. In fact, the trunk vertebrae of gopher snakes, Pituophis

(Fig. 1).

melanoleucus, were found to twist by up to 2.19 degrees and it

◦

was noted that torsion up to 2.89 was possible per vertebral joint

(Moon, 1999). 3.1. Cervical region

The cervical region is the most anterior region of the vertebral

3. Regionality in the axial skeleton column in tetrapods (Goodrich, 1958). The limit of the posterior

extension of the cervical region has been described as the anterior

In addition to variation in the morphology of individual verte- limit of hoxc6 expression (Burke et al., 1995). Anatomically, cervical

brae, there is tremendous diversity in the degree of vertebral vertebrae can be distinguished as having a foramen for the vertebral

72 A.B. Ward, R.S. Mehta / Zoology 117 (2014) 70–76

blood vessels (Liem et al., 2001). In amniotes, the ﬁrst two cervical 3.3. Sacral region

vertebrae are referred to as the atlas and axis, which are modiﬁed to

support and move the skull. Mammals have the most well-deﬁned True sacral vertebrae are only found in tetrapods. This region

cervical region, with a vast majority of mammals having seven cer- provides an important anchor for the pelvic girdle to transfer

vical vertebrae. The most well known exceptions to this rule are energy from the hindlimbs contacting the ground to the body axis.

sloths and manatees (Galis, 1999; see also Buchholtz, this issue). Frogs have a single sacral vertebra connected to a long urostyle,

Having seven cervical vertebrae does not constrain diversity in the which is derived from several fused postsacral vertebrae. Jorgensen

cervical region as vertebral shape can vary greatly in mammals. For and Reilly (2013) demonstrated that size of the sacral diapophysis,

example, giraffes have extremely long cervical vertebrae that grow which connects the sacrum to the elongated ilia, is a major pre-

disproportionately long throughout ontogeny (van Sittert et al., dictor of locomotory mode, even more so than relative length of

2010) and many cetaceans have compact and fused cervical verte- the hindlimbs. Birds have the greatest number of sacral vertebrae;

brae (Thewissen, 1988). Variation in cervical vertebral number is here, the lumbar and sacral vertebrae are fused into the synsacrum

more varied in other tetrapod groups, with Aves having the greatest (Gadow, 1933). The number of sacral vertebrae in mammals ranges

range in vertebral number for extant groups (Galis, 1999). While a from 2 to 9 (Narita and Kuratani, 2005).

large part of the avian vertebral column is fused, the cervical region

is not, permitting a high degree of ﬂexibility.

3.4. Caudal region

Although not often described, actinopterygian and chon-

drichthyan ﬁshes show modiﬁcations of the anterior vertebrae.

The caudal region is the most posterior region of the vertebral

Nowroozi et al. (2012) described the morphological differences

column. In ﬁshes, caudal vertebrae have fused hemal spines with

between the four anterior-most vertebrae and those in the

the exception of the ural vertebra. In sarcopterygians, the caudal

remaining precaudal region in striped bass (Morone saxatilis); the

region is posterior to the sacral vertebrae (Gadow, 1933). The cau-

ﬁrst four vertebrae have a stouter shape than the rest of the pre-

dal region varies dramatically in vertebral number, especially in

caudal vertebrae. In trumpet ﬁshes (Aulostomus sp.), the ﬁrst four

actinopterygian and chondrichthyan ﬁshes (Fig. 2). There is much

vertebrae are elongate and fused with the transverse processes,

variation in caudal vertebral number in Actinopterygii, ranging

forming a continuous shelf (Wheeler, 1955). Sygnathiformes also

from 9 in several members of the Tetraodontiformes (pufferﬁsh

have modiﬁcations of the anterior-most vertebrae. The ﬁrst 3–5 are

and their allies) to over 200 in electric eels (Electrophorus). While

highly elongated and fused (Pietsch, 1978). The Weberian appara-

the low end of the range is not present in Chondrichthyes, Alop-

tus in otophysan ﬁshes, a large group including approximately 8000

ias (thresher sharks) can have well over 200 caudal vertebrae

species, is a modification of the first five anterior vertebrae that

(Ward and Brainerd, 2007). Sarcopterygii also have a wide range

has often been suggested to be a major reason for the increased

in caudal vertebral number although the maximum number of

speciation of that group (Bird and Mabee, 2003; Nelson, 2006;

vertebrae does not come close to that of the former two lineages

Bird and Hernandez, 2009). Recently, Sallan (2012) described a

(Fig. 2).

tetrapodal-like cervical region for a fossil actinopterygian ﬁsh, Tar-

Of all of the axial regions, the caudal region has likely experi-

rasius. Claeson and Hilger (2011) have described modiﬁcations of

enced the greatest range of functional specializations; for example,

the anterior vertebral column in Squanitiformes, a lineage of chon-

the tail can be used for aquatic propulsion (both lateral and

drichthyan ﬁshes. Given the recent evidence of modiﬁcation of

dorsoventral), for prehensility, for support as an additional limb

anterior vertebrae, ﬁshes should be considered to have a cervical

or for balance in cursoriality, and for defense mechanisms (cau-

region just like the other groups of vertebrates.

dal autotomy). Prehensility is especially interesting as it is a

Overall, it is likely that the functional biology of the cervical

behavior that has evolved in almost every vertebrate class (Liem

region will be increasingly important to consider when attempting

et al., 2001). Investigating the caudal region in prehensile taxa

to understand regionality across Vertebrata.

would be particularly informative for understanding functional

convergence as well as phylogenetic constraints on vertebral mor-

phology.

3.2. Thoracolumbar region

4. Vertebral development

Technically, the thoracolumbar region is only present in mam-

mals, with the thoracic region separated from the lumbar region

The adult vertebra develops from the paraxial mesoderm.

by the presence of ribs (Romer, 1970). The vertebrae in this region

Early in development, the paraxial mesoderm segments through

have also been referred to as abdominal or dorsal. However, we con-

a process called somitogenesis. Somites separate from the unseg-

sider all non-cervical precaudal and presacral/precaudal vertebrae

mented paraxial mesoderm when genes from the Notch signaling

to be members of this region. This region has also been referred to

pathway are activated in a group of cells, resulting in epithelial-

as the dorsal or the abdominal region (Romer, 1970). In tetrapods,

ization of cells at the boundary of the segment. The periodicity

the vertebrae in this region lie between the pectoral and pelvic

of notch is partially set by its ability to activate an inhibitory

girdles, when present (Goodrich, 1958). In most ﬁshes, amphib-

protein (such as Mesp), which causes notch activity to cease.

ians, and lepidosaurs there is little distinction between the various

The inhibitory proteins are unstable and, once degraded, notch

postcervical and precaudal vertebrae (Romer, 1970). Within these

activity will continue (Gilbert, 2014). The period of this activ-

groups, the number of precaudal vertebrae can vary dramatically

ity and of inhibition appears to be set within a species although

(Fig. 2). In birds, the lumber vertebrae fuse with the sacral vertebrae

there is evidence that this period is not constant throughout

to form the synsacrum. Turtles fuse the thoracolumbar vertebrae

somitogenesis (e.g., Woltering et al., 2009; Gomez et al., 2008).

into the shell (Romer, 1970). Although lacking a complete carapace,

For example, the ﬁrst 5–6 somites in zebraﬁsh form every

the extinct parareptile Eunotosaurus demonstrates the transitions

20 min. The remaining somites form every 30 min (Kimmel et al.,

between typical slender ribs and the shell morphology of modern 1995).

turtles (Lyson et al., 2013). Most mammals have 19 thoracolumbar

Ultimately somite formation occurs due to an interaction

vertebrae (Narita and Kuratani, 2005), although their number can

between “clock” (notch) and “wavefront” (fgf) genes. The wavefront

range as high as 30.

genes are expressed in a gradient from posterior to anterior. The

A.B. Ward, R.S. Mehta / Zoology 117 (2014) 70–76 73

Chondrichthyes Actinopterygii Sarcopterygii Caudal Vertebral Number Caudal Vertebral

Precaudal Vertebral Number Chondrichthyes Actinopterygii Sarcopterygii Lissamphibia Squamata Aves Mammalia Caudal Vertebral Number Caudal Vertebral

Precaudal Vertebral Number Precaudal Vertebral Number Precaudal Vertebral Number Reconstructed Slopes (Caudal vs. Precaudal

Vertebral Number)

Fig. 2. Vertebral numbers were collected from the literature as well as from museum and personal collections. When a range was given, the maximal value was used.

Numbers were only included if the number of vertebrae in each region was given. The precaudal region was deﬁned as containing all vertebrae anterior to the caudal

vertebrae including the cervical, thoracic, lumbar, and sacral regions (Collar et al., 2013). The anterior border of the caudal region was deﬁned as the ﬁrst vertebra with fused

hemal spines (Actinopterygii, Chondrichthyes) or as the ﬁrst post-sacral vertebra (Sarcopterygii). References that were used for the vertebral numbers as well as the species

and vertebral numbers included can be found in Appendix A (supplementary online documents S1 and S2). The slopes from the regression analysis (caudal versus precaudal

vertebral numbers) were traced onto the phylogeny by minimizing the sum of squares using MacClade 4.06. Darker bars (black and blue) indicate a relatively higher number

of caudal vertebrae and lighter bars (white and yellow) indicate fewer caudal vertebrae in relation to precaudal vertebrae.

74 A.B. Ward, R.S. Mehta / Zoology 117 (2014) 70–76

clock is not functional in areas of high fgf concentration. As the tail 6. Evolution of vertebral number and axial patterning

bud extends posteriorly, cells in the anterior presomitic mesoderm

have a decreased fgf concentration and thus notch cycling can be Variation in vertebral number is one of the more striking dif-

activated (reviewed by Oates et al., 2012). ferences in the axial skeleton of vertebrates. While a handful of

The number of somites could be altered by changing either the studies have focused on the variation in vertebral numbers within

timing of the clock or the retreat speed of the wavefront. If the clock speciﬁc vertebrate clades (Ward and Brainerd, 2007; McDowall,

period increases, fewer somites will form, while each somite will be 2008; Mehta et al., 2010; Bergmann and Irschick, 2012), our interest

longer. The same effect would be caused by increasing the retreat here was to examine the diversity of vertebral numbers, speciﬁcally

speed of the wavefront (reviewed by Oates et al., 2012). However, the relationship between precaudal and caudal vertebral numbers

in both of these cases the overall length of the axial skeleton will not across Vertebrata. To do this, we collected regional vertebral num-

differ. Gomez et al. (2008) demonstrated that axial elongation may bers for over 1400 species of vertebrates. Vertebral numbers were

occur through a change in the rate of somitogenesis relative to the either mined from the literature or collected from specimens bor-

rate of overall development. Corn snakes, for example, increase the rowed from museums or private collections. Vertebral data were

number of vertebrae by increasing the rate of segmentation relative analyzed using reduced major axis (RMA) regression with variances

to the overall developmental rate (Gomez et al., 2008; Gomez and set as equal (JMP 8; SAS Institute, Cary, NC, USA). What we found

Pourquié, 2009). is that the relationship between precaudal and caudal vertebral

Vertebrae develop from the sclerotomal portion of the somite. numbers varies across the major clades. In all three groups (Chon-

Zebraﬁsh and amniotes go through a process of resegmentation drichthyes, Actinopterygii, and Sarcopterygii), there is a strong

by which the anterior part of the sclerotomal portion of a somite relationship between the numbers of precaudal and caudal verte-

will join with the posterior portion of the next anterior somite, brae, although the slopes differ (Fig. 2). In actinopterygians and

although zebraﬁsh are considered to have “leaky” resegmenta- chondrichthyans, the slope of the relationship between precau-

tion (Morin-Kensicki et al., 2002; Dequéant and Pourquié, 2008). dal and caudal vertebrae is >1, indicating that increasing overall

In anamniotes, the formation of vertebrae through resegmenta- vertebral number occurs primarily through the addition of cau-

tion is more controversial (Buckley et al., 2013). For example, dal vertebrae (95% conﬁdence interval of slope; Chondrichthyes:

Wake and Lawson (1973) suggested that since salamanders and 1.95–3.58, R = 0.54; Actinopterygii: 1.75–2.22, R = 0.50). In other

frogs have only a minimal sclerotome, resegmentation does not words, when vertebral number increases, it tends to increase in the

occur. caudal region. In Sarcopterygii, the slope was signiﬁcantly lower

than in the other two groups (95% conﬁdence interval: 0.07–0.12;

R = 0.27) indicating that increases in vertebral number tend to occur

5. Genetic basis for establishing regionality

in the precaudal region (Fig. 2).

Thus, actinopterygian and chondrichthyan ﬁshes tend to

As discussed, the axial skeleton has varying levels of regionality,

increase vertebral number in the caudal region whereas Sar-

with each region delineated by a change in morphology (Goodrich,

copterygii increase the number of precaudal vertebrae (Fig. 2).

1958; Fig. 1). The morphological differences that are associated

Based on our reconstruction, we hypothesize that the ancestral con-

with the different regions are thought to be due to hox expression

dition for vertebrates is to add caudal vertebrae, the character state

domains that are set up during early development (Burke et al.,

of the common ancestor of gnathostomes (Fig. 2). The pattern of

1995; Buchholtz, this issue). Despite the difference in the number

character state evolution is likely more complicated than what is

of cervical vertebrae between chicks and mice, Burke et al. (1995)

shown in Fig. 2, however, because the most basal group of living

showed that the anterior expression domain of hoxc6 always occurs

actinopterygian ﬁshes increases the number of precaudal verte-

at the transition between cervical and thoracic vertebrae.

brae (Polypteriformes; Ward and Brainerd, 2007). There are also

The gene hox10 is responsible for the transition between the

exceptions to this general trend within the sarcopterygian lineage;

rib-bearing thoracic region and the lumbar region, which con-

e.g., ichthyosaurs tend to have more caudal vertebrae (Buchholtz,

tains vertebrae that do not bear ribs (Romer, 1970; Wellik and

2001). Despite these exceptions, it is likely that different develop-

Capecchi, 2003). Despite this, in snakes hox10 is expressed in rib-

mental processes control the axial patterning of actinopterygian

bearing somites which are the result of a single base pair change

and sarcopterygian species. Within the actinopterygian lineage,

in the Hox binding site upstream of myf5, a gene required for rib

highly elongate body forms are probably due to changes in the

formation (Guerreiro et al., 2013). Interestingly, this same poly-

regulation of the total number of vertebrae, in particular, control

morphism is seen in mammalian species that are characterized

of axial or tail bud extension similar to the mechanism described

by longer ribcages due to additional thoracic segments (Guerreiro

for snakes (Gomez et al., 2008). In Sarcopterygii, there are likely

et al., 2013).

two different mechanisms, (i) a hox-derived mechanism that affects

The transition between precaudal and caudal vertebrae also

placement of the boundary that deﬁnes the precaudal region and

has a conserved expression pattern; this transition occurs at the

(ii) an axial elongation mechanism, which would affect the total

anterior expression domain of hoxd12 in chicks and mice as well

number of vertebrae (Gomez et al., 2008; Ward and Mehta, 2010).

as in zebraﬁsh (Burke et al., 1995; van der Hoeven et al., 1996).

What is unknown is whether ﬁshes also possess a mechanism to

In zebraﬁsh, there is strong expression of hoxd13 in the poste-

grow the relatively large number of precaudal vertebrae seen in

rior hindgut. This might be important for setting up the transition

snakes. The greatest number of actinopterygian precaudal verte-

between the precaudal and caudal regions by deﬁning the end of the

brae in this study was 134 in Callechelys melanotaenia, a member

gut tube (van der Hoeven et al., 1996). In the caecilian Ichthyophis

of the Anguilliformes or true eels. This is signiﬁcantly less than

kohtaoensis, hoxc13 is expressed at the boundary between the

the typical 200+ precaudal vertebrae seen in snakes. It is possi-

precaudal and caudal vertebrae (Woltering et al., 2009). It is cur-

ble that ﬁshes have lost the ability to grow the long precaudal

rently unknown whether any hox12 or hox13 genes are expressed

regions that are typical of snakes. Snipe eels (Anguilliformes) are

at the boundary between precaudal and caudal regions in chon-

the only clade reported to have over 600 vertebrae (Beebe and

drichthyans and lepidosaurs. Given the major distinction that

Crane, 1937). Snipe eels are extremely unusual as they are thought

occurs at this boundary, a better understanding of what controls

to add vertebrae throughout postnatal ontogeny. Whether these

this regional distinction will be critical for understanding axial pat-

vertebrae are added to the precaudal, caudal, or both regions, is terning. unknown.

A.B. Ward, R.S. Mehta / Zoology 117 (2014) 70–76 75

7. Future research directions and questions Buchholtz, E.A., 2001. Swimming styles in Jurassic ichthyosaurs. J. Vert. Paleo. 21,

61–73.

Buckley, D., Molnár, V., Németh, G., Petrneházy, Ö., Vörös, J., 2013.

As with many data sets, this one also brings up several possible .

‘Monster . .-omics’: on segmentation, re-segmentation, and vertebrae

questions that can lead to exciting new avenues of research. Here formation in amphibians and other vertebrates. Front. Zool. 10, 17,

http://dx.doi.org/10.1186/1742-9994-10-17.

are a few that we think should be considered:

Burke, A.C., Nelson, C.E., Morgan, B.A., Tabin, C., 1995. Hox genes and

the evolution of vertebrate axial morphology. Development 212, 333–

(i) While it is obvious that differential patterning occurs in 346.

Claeson, K.M., Hilger, A., 2011. Morphology of the anterior vertebral region in elas-

actinopterygian and sarcopterygian lineages, we have yet to

mobranchs: special focus, Squantiniformes. Foss. Rec. 14, 129–140.

understand the developmental control required for building a

Collar, D.C., Reynaga, C.M., Ward, A.B., Mehta, R.S., 2013. A revised met-

longer precaudal region. Do actinopterygian ﬁshes still main- ric for quantifying body shape in vertebrates. Zoology 116, 246–

257.

tain the ability to increase the precaudal region or has it been

Dequéant, M.-L., Pourquié, O., 2008. Segmental patterning of the vertebrate embry-

lost?

onic axis. Nat. Rev. Genet. 8, 370–382.

(ii) The caudal region is often ignored in studies of tetrapod groups, Ellenberger, W., Baum, H., Dittrich, H., Brown, L.S., 1956. An Atlas of Animal Anatomy

for Artists. Dover Publications, New York.

in part due to the likelihood of damage to the posterior end

Gadow, H.F., 1933. The Evolution of the Vertebral Column. Cambridge University

of specimens. It is possible that there is greater range in

Press, London.

the number of caudal vertebrae in sarcopterygians than has Galis, F., 1999. Why do almost all mammals have seven cervical vertebrae? Devel-

opmental constraints, Hox genes, and cancer. J. Exp. Zool. 285, 19–26.

been reported here. What is the relationship between caudal

Gilbert, S.F., 2014. Developmental Biology. Sinauer Associates, Inc., Sunderland, MA.

vertebral number and functional specializations? Is there mor-

Gomez, C., Pourquié, O., 2009. Developmental control of segment numbers in ver-

phological convergence in the caudal region during repeated tebrates. J. Exp. Zool. 312B, 533–544.

functional specializations across vertebrates? Gomez, C., Özbudak, E.M., Wunderlich, J., Baumann, D., Lewis, J., Pourquié, O.,

2008. Control of somite number in vertebrate embryos. Nature 454, 335–

(iii) While we primarily discussed the broad patterns of axial

339.

regionalization in the different groups of vertebrates, there are

Goodrich, E.S., 1958. Studies on the Structure and Development of Vertebrates, vol.

exceptions such as in the squamate Delma fraseri (Pygopodi- 1. Dover Publications, New York.

Guerreiro, I., Nunes, A., Woltering, J.M., Casaca, A., Nóvoa, A., Vinagre, T., Hunter, M.E.,

dae) that has a relatively large number of caudal vertebrae

Duboule, D., Mallo, M., 2013. Role of a polymorphism in a Hox/Pax-responsive

relative to the precaudal vertebrae. Additionally, ichthyosaurs

enhancer in the evolution of the vertebrate spine. Proc. Natl. Acad. Nat. Sci. U. S.

also have relatively large numbers of caudal vertebrae A. 110, 10682–10686.

Janvier, P., 1997. Vertebrata. Animals with Backbones. Version 01 January 1997

(Buchholtz, 2001). Which selective pressures have led to these

(Under Construction), http://tolweb.org/Vertebrata/14829/1997.01.01 in The

exceptions?

Tree of Life Web Project, http://tolweb.org/

(iv) When an earlier version of Fig. 2 was initially published by Jordan, D.S., 1891. Relations of temperature to vertebrae among ﬁshes. Proc. U. S.

Nat. Mus. 14, 107–120.

Ward and Brainerd (2007), there was unﬁlled morphospace in

Jorgensen, M.E., Reilly, S.M., 2013. Phylogenetic patterns of skeletal morphometrics

the lower right quadrant, which is now ﬁlled by Sarcopterygii.

and pelvic traits in relation to locomotor mode in frogs. J. Evol. Biol. 26, 929–

The only remaining empty morphospace is in the upper right 943.

Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T., 1995. Stages of

quadrant. Are there species that can be found in this area? If

embryonic development of the zebraﬁsh. Dev. Dyn. 203, 253–310.

not, why are there no species in this region of morphospace; is

Liem, K.F., Bemis, W.E., Walker Jr., W.F., Grande, L., 2001. Functional Anatomy of the

this due to developmental constraints or to a lack of selection Vertebrates: An Evolutionary Perspective. Harcourt College Publishers, Philadel-

for equally large numbers of precaudal and caudal vertebrae? phia.

Lindsey, C.C., 1975. Pleomerism, widespread tendency among related ﬁsh species

for vertebral number to be correlated with maximum body length. J. Fish. Res.

Acknowledgements Board Can. 32, 2453–2469.

Lyson, T.R., Beaver, G.S., Scheyer, T.M., Hsiang, A.Y., Gauthier, J.A., 2013. Evolutionary

origin of the turtle shell. Curr. Biol. 23, 1–7.

We would like to thank the symposium organizers, John Long

McDowall, R.M., 2008. Jordan’s and other ecogeographical rules, and the vertebral

and Nadja Schilling for the invitation to join this volume. We number in ﬁshes. J. Biogeogr. 35, 501–508.

Mehta, R.S., Ward, A.B., Alfaro, M.E., Wainwright, P.C., 2010. Body elongation in eels.

would also like to thank Vikram Baliga, Danielle Pruitt, Crystal Rey-

Integr. Comp. Biol. 50, 1091–1105.

naga, and Michaela Tondi who aided in gathering vertebral data.

Moon, B.R., 1999. Testing an inference of function from structure: snake vertebrae

This manuscript was improved through discussions with David do the twist. J. Morphol. 241, 217–225.

Morin-Kensicki, E.M., Melancon, E., Eisen, J.S., 2002. Segmental relationship

Wake and Nathan Kley as well as comments from two anonymous

between somites and vertebral column in zebraﬁsh. Development 129, 3851–

reviewers. The authors were supported by funds from the Biology

3860.

Department at Adelphi University and NSF grants IOS 0819009 and Mosauer, W., 1932. On the locomotion of snakes. Science 76, 583–585.

Narita, Y., Kuratani, S., 2005. Evolution of the vertebral formulae in mammals: a

REU 1126349 to R.S.M.

perspective on developmental constraints. J. Exp. Zool. 304 B, 1–16.

Nelson, G.J., 2006. Fishes of the World, 4th ed. John Wiley & Sons, Inc., Hoboken.

Appendix A. Supplementary data Nowroozi, B.N., Harper, C.J., De Kegel, B., Adriaens, D., Brainerd, E.L., 2012. Regional

variation in morphology of vertebral centra and intervertebral joints in striped

bass, Morone saxatilis. J. Morphol. 273, 441–452.

Supplementary material related to this article can be found, in

Oates, A.C., Morelli, L.G., Ares, S., 2012. Patterning embryos with oscillations: struc-

the online version, at http://dx.doi.org/10.1016/j.zool.2013.10.006. ture, function, and dynamics of the vertebrate segmentation clock. Development

139, 625–639.

Parker, T.J., 1900. A Manual of Zoology. MacMillan Co., New York.

References Pietsch, T.W., 1978. Evolutionary relationships of the sea moths (Teleostei: Pegasi-

dae) with a classiﬁcation of gasterosteiform families. Copeia 1978, 517–529.

Romer, A.S., 1966. Vertebrate Paleontology, 3rd ed. University of Chicago Press,

Arratia, G., Schultze, H.-P., Casciotta, J., 2001. Vertebral column and associated

Chicago.

elements in dipnoans and comparisons with other ﬁshes: development and

Romer, A.S., 1970. The Vertebrate Body, 4th ed. Saunders, Philadelphia.

homology. J. Morphol. 250, 101–172.

Sallan, L.C., 2012. Tetrapod-like axial regionalization in an early ray-ﬁnned ﬁsh. Proc.

Beebe, W., Crane, J., 1937. Deep-sea ﬁshes of the Bermuda oceanographic expedi-

R. Soc. Lond. 279, 3264–3271.

tions. Family Nemichthyidae. Zoologica 22, 249–383.

Schultze, H.-P., Arratia, G., 1988. Reevaluation of the caudal skeleton of some

Bergmann, P.J., Irschick, D.J., 2012. Vertebral evolution and diversiﬁcation of squa-

actinopterygian ﬁshes. II. Hiodon, Elops and Albula. J. Morphol. 195, 257–303.

mate reptiles. Evolution 66, 1044–1058.

Thewissen, J.G.M. (Ed.), 1988. The Emergence of Whales: Evolutionary Patterns in

Bird, N.C., Hernandez, L.P., 2009. Building an evolutionary innovation: differen-

the Origin of Cetacea. Plenum, New York.

tial growth in the modiﬁed vertebral elements of the zebraﬁsh. Zoology 112,

97–112. Tytell, E.D., Borazjani, I., Sotiropoulos, F., Baker, T.V., Andersos, E.J., Lauder, G.V., 2010.

Disentangling the functional roles of morphology and motion in the swimming

Bird, N.C., Mabee, P.M., 2003. Developmental morphology of the axial skeleton of

of ﬁsh. Integr. Comp. Biol. 50, 1140–1154.

the zebraﬁsh, Danio rerio (Ostariophysi: Cyprinidae). Dev. Dyn. 228, 337–357.

76 A.B. Ward, R.S. Mehta / Zoology 117 (2014) 70–76

van der Hoeven, F., Sordino, P., Fraudeau, N., Izpisúa-Belmonte, J.-C., Duboule, D., Ward, A.B., Mehta, R.S., 2010. Axial elongation in ﬁshes: using morphological

1996. Teleost HoxD and HoxA genes: a comparison with tetrapods and func- approaches to elucidate developmental mechanisms in studying body shape.

tional evolution of the HOXD complex. Mech. Dev. 54, 9–21. Integr. Comp. Biol. 50, 1106–1119.

van Sittert, S.J., Skinner, J.D., Mitchell, G., 2010. From fetus to adult – an Wellik, D.M., Capecchi, M.R., 2003. Hox10 and Hox11 genes are required to globally

allometric analysis of the giraffe vertebral column. J. Exp. Zool. 314 B, pattern the mammalian skeleton. Science 301, 363–367.

469–479. Wheeler, A.C., 1955. A preliminary revision of the ﬁshes of the genus Aulostomus. J.

Wake, D.B., Lawson, R., 1973. Developmental and adult morphology of the vertebral Nat. Hist. 8 (12), 613–623.

column in the plethodontid salamander Eurycean bislineata, with comments on Woltering, J.M., Vonk, F.J., Müller, H., Bardine, N., Tuduce, I.L., de Bakker, M.A.G.,

vertebral evolution in the Amphibia. J. Morphol. 139, 251–300. Knöchel, W., Sirbu, I.O., Durston, A.J., Richardson, M.K., 2009. Axial patterning

Ward, A.B., Brainerd, E.L., 2007. Evolution of axial patterning in elongate ﬁshes. Biol. in snakes and caecilians: evidence for an alternative interpretation of the Hox

J. Linn. Soc. 90, 97–116. code. Dev. Biol. 332, 82–89.